Surgical Resection Is Still Better Than Endoscopic Resection For Patients With 2-5 Cm Gastric Gastrointestinal Stromal Tumors-A Propensity Score Matching Analysis

BackgroundGastrointestinal stromal tumor(GIST)is the most common mesenchymal tumor of the digestive tract.GISTs can arise anywhere in the gastrointestinal tract,most commonly in the stomach.GISTs are currently considered to have malignant potential,but most gastric stromal tumors are indolent.According to current guidelines,patients with gastric GISTs smaller than 2 cm and without high-risk features should receive regular examinations.For patients with primary local gastric GISTs larger than 2 cm,surgical resection(SR)is the preferred treatment.With the continuous advancement of medical devices and technologies in recent years,endoscopic resection(ER)has been increasingly used in the treatment of gastric stromal tumors due to its its minimal invasiveness and rapid recovery.However,the difficulty in guaranteeing en bloc resection,resulting in a potential risk of recurrence,is an ongoing concern for clinicians.Therefore,there is still considerable controversy between surgeons and physicians in the treatment and management of 2-5 cm gastric stromal tumors.Many studies of endoscopic resection for gastric GISTs have shown short-term safety,but additional evidence is still needed to confirm the long-term efficacy of endoscopic resection for stromal tumors.In addition,most studies included gastric GISTs less than 2 cm in diameter,and few studies compared endoscopic resection with surgical resection for gastric GISTs accurate to the 2-5 cm range.In previously published studies,the number of patients and tumor diameters in the surgical group were often much larger than those in the endoscopic group,the proportion of patients was disproportionate,and the baseline characteristics of tumors were significantly different,resulting in a large selection bias.In this study,we established baselinecomparable cohorts by propensity score matching to investigate the safety and efficacy of endoscopic versus surgical resection for the treatment of 2-to 5-cm gastric GISTs.MethodsThis study screened patients with 2-5 cm gastric stromal tumors diagnosed between 2010 and 2020 who underwent endoscopic and surgical resection.The overall cohort initially included 282 patients.Oncological outcomes of 52 pairs of patients from the endoscopic and surgical resection groups were compared by 1:1 propensity score matching for age,sex,BMI,tumor size,tumor location,growth type,and mitotic variables.At the same time,the baseline characteristics of all the included patients were described and compared,and the case information of patients requiring special attention was listed,and the subgroup analysis of the overall cohort was also performed.Categorical variables were analyzed using Pearson’s chi-square test or Fisher’s exact test.Continuous variables were compared using the Mann-Whitney U test.KaplanMeier method and Log-rank test were performed separately for survival analysis and to assess differences in survival time.Statistical significance was considered when P<0.05.ResultsIn terms of the overall cohort,before matching,tumor location(P=0.024),tumor size(P<0.001),mitotic index(P=0.048),modified NIH risk category(P<0.001),and tumor growth type(P=0.038)were statistically different.No significant differences were observed at baseline between the two groups after matching.Before matching,operation time(P<0.001),en bloc resection(P<0.001),estimated blood loss(P=0.037),liquid diet time(P=0.001),postoperative hospital stays(P<0.001),and postoperative adverse events(P=0.001)were statistically different.After matching,the endoscopy group was not significantly superior in terms of estimated blood loss(P=0.741)and liquid diet time(P=0.055).Adverse events such as secondary surgery occurred in 6 patients(11.5%)in the endoscopic resection group,and no postoperative stenosis,leakage and other adverse events occurred in the surgical resection group.The R0 resection rate was similar between the two groups,but 11 patients(21.1%)in the endoscopic resection group failed to achieve en bloc resection.After matching,the overall mean follow-up time was 1781 days in the endoscopic resection group and 1960 days in the surgical resection group.There was no statistical difference in progressionfree survival between the two groups.However,it cannot be ignored that 5 patients experienced excessive bleeding or difficulty in resection during endoscopic treatment and were converted to surgery.In the subgroup of surgical versus endoscopic resection of 2-3 cm gastric GISTs,before matching,operation time(P=0.034),en bloc resection(P=0.006),liquid diet time(P=0.049),postoperative hospital stays(P=0.018)and adverse events(P=0.006)were significantly different.After matching,there were no statistically significant differences in operative time(P=0.062)and adverse events(P=0.056).In the laparoscopic versus endoscopic resection subgroup for those 2-5 cm gastric GISTs originating from the muscularis propria,before matching,there were significant differences in operation time(P=0.001),en bloc resection(P=0.002),liquid diet time(P=0.008),postoperative hospital stays(P=0.001),and adverse events(P=0.010).However,after matching,there was no significant difference between the two groups in terms of liquid diet time(P=0.072)and adverse events(P=0.237).A subgroup analysis was subsequently attempted,comparing surgical versus endoscopic resection of 3-5 cm gastric GISTs.Before matching,statistical differences were only observed in en bloc resections(P<0.001).After matching,there were significant differences in operative time(P=0.011)and en bloc resection(P=0.005).However,there are few patients in this subgroup and the statistical power is low,and further studies with larger sample size are needed.ConclusionER might be comparable with SR for treating 2-3 cm gastric GISTs.SR is still considered the standard treatment for 3-5 cm gastric GISTs,while ER should be recorded in detail and under close surveillance.Surgical resection is recommended if the tumor has high mitotic index or mucosal ulceration.All patients should be carefully evaluated to select the most appropriate resection procedure.Multi-center random control trials are warranted to settle this argument in the future.